Introduction

In 1983 the United Kingdom was very clearly a unitary state. Despite consisting of three distinct nations and a territory whose national status is disputed, all legislative power and executive responsibility lay with Westminster and Whitehall. Proposals to introduce separately elected devolved assemblies in Scotland and Wales had failed to secure adequate support in referendums held in 1979. Meanwhile in Northern Ireland, where a devolved parliament and government had existed until 1972, hopes of reintroducing devolution were looking unlikely to be fulfilled in the wake of continuing civil strife. All that was in place was an assembly that had the ability to scrutinise the Westminster government's Secretary of State, but which was being boycotted by its nationalist members - and before long even that limited institution was to come to an unheralded end.

Thirty years on the position is very different. Both Scotland and Wales now have separately elected representative bodies that have the ability to make laws without reference to Westminster, and governments that can decide for themselves how many of the public services in their country should be funded and run. Northern Ireland too now enjoys devolution once more, albeit in a very different form to what was in place before 1972 or is to be found anywhere else in the UK today. In order to ensure that power is shared between the two distinct ethno-religious communities living there - a Protestant community whose roots lie in 17th-century immigration from Scotland and a Catholic community that shares many ties and affinities with the rest of the island of Ireland - not only is the territory's law-making Assembly elected by proportional representation, but ministerial posts are allocated in that way too.

Much of the impetus for these developments lay in a wish to demonstrate that the United Kingdom was capable of accommodating the diverse identities and aspirations that lie within it, and thereby help end disputes that threatened its territorial integrity (Aughey, 2001; Bogdanor, 1999; Mackintosh, 1998). It was hoped that by giving Wales and (especially) Scotland their own separate political institutions that had the ability to determine much of their country's domestic affairs, demands for independence that were being spearheaded by nationalist political parties would be headed off. Meanwhile in Northern Ireland it was anticipated that, together with the creation of institutions to facilitate dialogue and collaboration with the neighbouring Irish Republic, giving nationalist politicians who represented the minority Catholic population a guaranteed role in the territory's government would facilitate accommodation of the identities and aspirations of Catholics, while recognising that, for the time being at least, a majority of the territory's population wanted to remain part of the UK (McGarry and O'Leary, 1995; 2006).

These innovations were not, however, without their critics. In Scotland in particular, it was argued that the creation of a separate parliament would put the country on a 'slippery slope' towards independence (Dalyell, 1977). The new institution's politicians would inevitably want to increase their power at Westminster's expense. Creating distinctively Scottish political symbols would simply stimulate and fuel Scottish national identity rather than strengthen adherence to Britishness and the UK state (Thatcher, 1998). At the same time, relations with England would be soured because while Scotland and Wales would now be able settle many of their own affairs for themselves, politicians from the rest of the UK could still meddle in the affairs of England, where no devolution was in place and where public spending per head would continue to be less than that enjoyed elsewhere. It was even suggested that an 'English backlash' might ensue (Wright, 2000).

In Northern Ireland there were doubts whether the devolution settlement that was agreed in talks concluded on Good Friday 1998 would succeed in bringing all parts of the territory together. One of the two main Protestant political parties, the Democratic Unionist Party (DUP), chose not to be party to the Agreement. For many unionist politicians, much rested on whether the proposals for decommissioning weaponry that had been accumulated during 30 years of civil strife, not least by the Provisional Irish Republican Army (IRA), would be seen to work in practice. Meanwhile, by providing guaranteed representation to both unionist and nationalist politicians in the executive[1] as well as the legislature, and in requiring that major legislative decisions have the support of the representatives of both communities in the Assembly, some commentators were concerned that the settlement would simply reinforce existing lines of division as well as potentially undermine the effectiveness and accountability of ministerial decision-making (Wilson and Wilford, 2003; Horowitz, 2001).

Initially, at least, it was the doubts about the stability and effectiveness of the settlement in Northern Ireland that appeared to have the greater force. The Assembly only operated for 10 weeks before it had to be suspended because of a perceived lack of progress on arms decommissioning. After a number of other hiccups the devolved institutions actually lay in abeyance for nearly five years, from 2002 to 2007. However, some nine years after the conclusion of the Good Friday Agreement (also known as the Belfast Agreement) and with the issue of decommissioning finally resolved, the devolved institutions began to operate in 2007 on a continuous basis, and this time with the DUP as full participants. Indeed Northern Ireland was faced with the remarkable sight of the DUP's leader, the Revd. Ian Paisley, a long-standing and rhetorically flamboyant advocate of hard line unionism, working in government with Martin McGuinness, who had formerly been an active member of the Provisional IRA.

More recently, however, it has been the durability of Scotland's constitutional position that has looked to be in most doubt. Far from killing nationalism 'stone dead' as had been anticipated by the former Labour Cabinet Minister, (Lord) George Robertson (Hassan, 2011), it soon became apparent that elections to the Scottish Parliament were proving to be a relatively happy hunting ground for the Scottish National Party (SNP) (as indeed were Assembly elections for the Welsh nationalist party, Plaid Cymru, in Wales). In 2007 the SNP narrowly succeeded in securing the largest number of MSPs, and for the next four years formed a minority government. Still, the party's ambitions to hold a referendum on whether Scotland should leave the UK and become an independent country remained on hold thanks to the absence of a majority for such a move in the Edinburgh Parliament. However, in 2011 the nationalists dramatically won an overall majority and the UK government accepted that this gave the party the moral right to hold an independence referendum. After extended negotiations between the two governments the way has now been cleared for a referendum on independence to be held in September 2014. Should there be a majority 'Yes' vote then preparations will begin to be made for Scotland to leave the UK.

In this chapter we look underneath the bonnet of these election outcomes and political developments to examine the trends in identities and constitutional preferences among the general public since the advent of devolution in Scotland and Northern Ireland.[2] Has public support for remaining part of the UK in fact either grown or diminished in the wake of constitutional change? At the same time we also examine developments in the UK's largest component: England. Does it accept the patchwork of asymmetric devolution that has emerged or are there signs of discontent? To answer these questions we address three key issues that are central to understanding the role that devolution has so far played in helping to maintain - or fracture - the UK.

First, what has happened to patterns of national identity? A shared sense of Britishness is often regarded as the emotional glue that helps keep the UK together. Yet it coexists alongside other 'national' identities, Scottish, English, Irish, Northern Irish and Welsh. Has the establishment of a distinctive Scottish Parliament undermined adherence to Britishness north of the border? Has England become more aware of its own separate English identity? And has greater acknowledgement of the aspirations of many Northern Irish Catholics served in practice to reinforce in them a sense of having a distinct Irish identity?

Even if the pattern of national identity has not changed, people's preferences as to how they would like to be governed may have done so. The electoral success of the SNP certainly gives good reason to wonder whether in Scotland support for leaving the UK has increased since devolution has been in place. Meanwhile, perhaps England has begun to question whether it should continue to accommodate the demands of its seemingly increasingly truculent neighbour and/or whether it should be enjoying some form of devolution itself. At the same time, we might wonder now that the devolved institutions in Northern Ireland have been up and running for a while, whether more people in the region are willing to accept its continued role as part of the UK. So, our second key issue is, 'What trends have emerged in how people in each part of the UK would like to be governed?'

Finally, we turn to the relationship between identity and preferences, and ask how far those with different national identities disagree about how their part of the UK should be run. If devolution really has been successful in strengthening public support for keeping the UK in its current form, then arguably those whose primary sense of identity is something other than British should have come to accept that where they live should remain in the UK. In short it should have succeeded in healing the constitutional divisions between those with different identities. On the other hand, if the fears of those who were critical of devolution have been realised we might find those who do not feel primarily British have become even less likely to accept that they should be part of the British state. Those who feel Scottish or Irish may have become even more determined to want to leave the UK, while those who primarily feel English may have particularly come to wonder whether their part of the UK is getting a raw deal. So our third issue is then, 'Has the link between identity and constitutional preference strengthened or weakened?'

To answer these questions, we draw on data from three complementary sources. The first is the British Social Attitudes survey which, since the late 1990s, has been asking its respondents in England, on a regular basis, about both their national identity and their constitutional preferences. In addition, since 1983 British Social Attitudes has asked all respondents in Britain what they think the constitutional position of Northern Ireland should be, and so we can examine how far the rest of the UK believes that territory should form part of the British state.

British Social Attitudes covers Scotland too, but it contains too few respondents in any one year to provide us with reliable estimates of the distribution of public opinion north of the border. Instead for this information we turn to British Social Attitudes' sister survey, the Scottish Social Attitudes survey, which was first established in the immediate wake of the creation of the Scottish Parliament in 1999 and is conducted using the same methodological approach as British Social Attitudes. It is the only survey to have asked the same questions about national identity and constitutional preferences in a consistent manner ever since the advent of devolution in 1999.

British Social Attitudes has never been conducted in Northern Ireland, but in 1989 NatCen Social Research helped establish an equivalent survey there, known as the Northern Ireland Social Attitudes survey. Funding ended in 1996, but two years later researchers at Queen's University, Belfast and the University of Ulster launched the Northern Ireland Life and Times survey, using much the same methodological approach as the Northern Ireland Social Attitudes survey and, in turn, British Social Attitudes. Between them the Northern Ireland Social Attitudes survey and the Northern Ireland Life and Times survey provide the longest running series of Northern Irish data on our subject matter. Furthermore, the similarity in the way in which all three surveys are conducted means they provide the most robust opportunities for comparing trends in opinion across all three parts of the UK.[3]